@cubis/foundry 0.3.83 → 0.3.84

package/workflows/workflows/agent-environment-setup/platforms/gemini/skills/frontend-design-screen-brief/SKILL.md

@@ -0,0 +1,45 @@
+---
+name: frontend-design-screen-brief
+description: Turn canonical design state plus overlays into a compact, high-signal screen brief for Stitch, frontend implementation, or Flutter/mobile work.
+---
+# Frontend Design Screen Brief
+
+## Purpose
+
+Produce the compact screen-level brief that downstream tools or engineers can execute without losing the product's visual and interaction intent.
+
+## When to Use
+
+- Preparing a Stitch generation or edit prompt
+- Creating a Flutter/mobile screen implementation brief
+- Turning a broad design system into a concrete page or flow
+
+## Instructions
+
+1. **Resolve the design state first** — Read `docs/foundation/DESIGN.md` and any relevant overlay before writing the brief.
+2. **Describe behavior, not just appearance** — Include hierarchy, interaction states, content priority, empty states, and success/failure feedback.
+3. **Keep the prompt compact** — The screen brief should be concise enough for a remote service like Stitch and concrete enough for a human implementer.
+4. **Choose platform-specific emphasis** — For Flutter/mobile, emphasize thumb reach, section staging, safe areas, and navigation patterns. For web, emphasize layout elasticity and information density.
+5. **Write acceptance constraints** — Explicitly state what must be preserved and what generic shortcuts are not acceptable.
+
+## Output Format
+
+Deliver:
+
+1. Screen goal
+2. Hierarchy and layout summary
+3. Token and component cues
+4. Interaction/motion notes
+5. Anti-slop constraints
+
+## Gemini Platform Notes
+
+- Workflow and agent routes are compiled into `.gemini/commands/*.toml` TOML command files.
+- Commands use `{{args}}` for user input, `!{shell command}` for shell output, `@{file}` for file content.
+- Specialists are internal postures (modes of reasoning), not spawned subagent processes.
+- Gemini does not support `context: fork` — all skill execution is inline within the current session.
+- Skills are loaded via MCP when the Cubis Foundry MCP server is configured. Local `.agents/skills/` paths serve as hints.
+- User arguments are passed as natural language in the activation prompt.
+- Rules file relative to the mirrored skill directory: `../../rules/GEMINI.md`.
+- Reference files are loaded relative to the skill directory under `.agents/skills/<skill-id>/`.
+- MCP skill tools (`skill_search`, `skill_get`, `skill_validate`, `skill_get_reference`) are available when MCP is connected.

package/workflows/workflows/agent-environment-setup/platforms/gemini/skills/frontend-design-style-selector/SKILL.md

@@ -0,0 +1,44 @@
+---
+name: frontend-design-style-selector
+description: Select a non-generic visual direction from Foundry's normalized design datasets and align it to the product's domain, audience, and platform.
+---
+# Frontend Design Style Selector
+
+## Purpose
+
+Choose the visual direction before any screen or component work starts so the result has a deliberate point of view instead of a generic app-template look.
+
+## When to Use
+
+- Starting a new design system or screen family
+- Improving a weak or generic UI prompt before Stitch generation
+- Choosing a consistent mood for a Flutter/mobile app
+
+## Instructions
+
+1. **Use only Foundry datasets** — Load `workflows/design-datasets/style-directions.json`, `layout-patterns.json`, and `component-motifs.json` first.
+2. **Select one primary direction and one supporting motif** — Avoid mixing many vibes. State the lead direction, supporting motif, and what should be excluded.
+3. **Tie the direction to the product domain** — State why the chosen direction fits the product audience, task density, and emotional tone.
+4. **Write anti-slop constraints** — Explicitly call out the visual clichés to avoid, such as generic wellness gradients, default dashboard cards, or unstructured glassmorphism.
+5. **Feed the result into `frontend-design-system` or `frontend-design-screen-brief`** — This skill chooses direction; it does not finish the design state by itself.
+
+## Output Format
+
+Deliver:
+
+1. Primary style direction
+2. Supporting motif
+3. Excluded clichés
+4. Typography, palette, shape, and motion intent summary
+
+## Gemini Platform Notes
+
+- Workflow and agent routes are compiled into `.gemini/commands/*.toml` TOML command files.
+- Commands use `{{args}}` for user input, `!{shell command}` for shell output, `@{file}` for file content.
+- Specialists are internal postures (modes of reasoning), not spawned subagent processes.
+- Gemini does not support `context: fork` — all skill execution is inline within the current session.
+- Skills are loaded via MCP when the Cubis Foundry MCP server is configured. Local `.agents/skills/` paths serve as hints.
+- User arguments are passed as natural language in the activation prompt.
+- Rules file relative to the mirrored skill directory: `../../rules/GEMINI.md`.
+- Reference files are loaded relative to the skill directory under `.agents/skills/<skill-id>/`.
+- MCP skill tools (`skill_search`, `skill_get`, `skill_validate`, `skill_get_reference`) are available when MCP is connected.

package/workflows/workflows/agent-environment-setup/platforms/gemini/skills/frontend-design-system/SKILL.md

@@ -0,0 +1,46 @@
+---
+name: frontend-design-system
+description: Build or refresh the canonical design-system state for a repo, including DESIGN.md, overlays, token language, and platform-specific adaptation notes.
+---
+# Frontend Design System
+
+## Purpose
+
+Author and refresh the project's canonical design-system state so design, Stitch generation, and implementation share the same visual language.
+
+## When to Use
+
+- `docs/foundation/DESIGN.md` is missing or stale
+- A new design direction needs to be made durable
+- Multi-screen work needs consistent tokens and component vocabulary
+- A Flutter/mobile app needs a stable design foundation before implementation
+
+## Instructions
+
+1. **Create or refresh `docs/foundation/DESIGN.md`** — This file is the source of truth for visual direction, token language, component vocabulary, and motion rules.
+2. **Write scoped overlays only when needed** — Use the page overlays directory under `docs/foundation/design/pages/`, the flow overlays directory under `docs/foundation/design/flows/`, or the mobile overlays directory under `docs/foundation/design/mobile/` when a narrower slice of the product needs additional detail.
+3. **Use semantic token naming** — Load `workflows/design-datasets/token-language.json` and preserve semantic roles instead of raw color dumps.
+4. **Preserve source metadata in the authoring process, not the final doc body** — The final design docs should be crisp and implementation-oriented, but the chosen direction should come from normalized datasets with source provenance.
+5. **Mirror to `.stitch/DESIGN.md` when Stitch flows are in scope** — The mirror should be generated from the canonical state and applicable overlays, never hand-maintained separately.
+
+## Output Format
+
+Deliver:
+
+1. Canonical design-system summary
+2. Token language
+3. Component vocabulary
+4. Overlay files created or refreshed
+5. Whether `.stitch/DESIGN.md` was updated
+
+## Gemini Platform Notes
+
+- Workflow and agent routes are compiled into `.gemini/commands/*.toml` TOML command files.
+- Commands use `{{args}}` for user input, `!{shell command}` for shell output, `@{file}` for file content.
+- Specialists are internal postures (modes of reasoning), not spawned subagent processes.
+- Gemini does not support `context: fork` — all skill execution is inline within the current session.
+- Skills are loaded via MCP when the Cubis Foundry MCP server is configured. Local `.agents/skills/` paths serve as hints.
+- User arguments are passed as natural language in the activation prompt.
+- Rules file relative to the mirrored skill directory: `../../rules/GEMINI.md`.
+- Reference files are loaded relative to the skill directory under `.agents/skills/<skill-id>/`.
+- MCP skill tools (`skill_search`, `skill_get`, `skill_validate`, `skill_get_reference`) are available when MCP is connected.

package/workflows/workflows/agent-environment-setup/platforms/gemini/skills/git-workflow/SKILL.md

@@ -67,6 +67,6 @@ When advising on git workflows, provide:
 - Gemini does not support `context: fork` — all skill execution is inline within the current session.
 - Skills are loaded via MCP when the Cubis Foundry MCP server is configured. Local `.agents/skills/` paths serve as hints.
 - User arguments are passed as natural language in the activation prompt.
-- Rules file: `.gemini/GEMINI.md`.
+- Rules file relative to the mirrored skill directory: `../../rules/GEMINI.md`.
 - Reference files are loaded relative to the skill directory under `.agents/skills/<skill-id>/`.
 - MCP skill tools (`skill_search`, `skill_get`, `skill_validate`, `skill_get_reference`) are available when MCP is connected.

package/workflows/workflows/agent-environment-setup/platforms/gemini/skills/golang-best-practices/SKILL.md

@@ -78,6 +78,6 @@ No helper scripts are required for this skill right now. Keep execution in `SKIL
 - Gemini does not support `context: fork` — all skill execution is inline within the current session.
 - Skills are loaded via MCP when the Cubis Foundry MCP server is configured. Local `.agents/skills/` paths serve as hints.
 - User arguments are passed as natural language in the activation prompt.
-- Rules file: `.gemini/GEMINI.md`.
+- Rules file relative to the mirrored skill directory: `../../rules/GEMINI.md`.
 - Reference files are loaded relative to the skill directory under `.agents/skills/<skill-id>/`.
 - MCP skill tools (`skill_search`, `skill_get`, `skill_validate`, `skill_get_reference`) are available when MCP is connected.

package/workflows/workflows/agent-environment-setup/platforms/gemini/skills/integration-testing/SKILL.md

@@ -104,6 +104,6 @@ Provide a structured methodology for writing integration tests that verify corre
 - Gemini does not support `context: fork` — all skill execution is inline within the current session.
 - Skills are loaded via MCP when the Cubis Foundry MCP server is configured. Local `.agents/skills/` paths serve as hints.
 - User arguments are passed as natural language in the activation prompt.
-- Rules file: `.gemini/GEMINI.md`.
+- Rules file relative to the mirrored skill directory: `../../rules/GEMINI.md`.
 - Reference files are loaded relative to the skill directory under `.agents/skills/<skill-id>/`.
 - MCP skill tools (`skill_search`, `skill_get`, `skill_validate`, `skill_get_reference`) are available when MCP is connected.

package/workflows/workflows/agent-environment-setup/platforms/gemini/skills/java-best-practices/SKILL.md

@@ -73,6 +73,6 @@ Produces Java code using records, sealed types, pattern matching switch expressi
 - Gemini does not support `context: fork` — all skill execution is inline within the current session.
 - Skills are loaded via MCP when the Cubis Foundry MCP server is configured. Local `.agents/skills/` paths serve as hints.
 - User arguments are passed as natural language in the activation prompt.
-- Rules file: `.gemini/GEMINI.md`.
+- Rules file relative to the mirrored skill directory: `../../rules/GEMINI.md`.
 - Reference files are loaded relative to the skill directory under `.agents/skills/<skill-id>/`.
 - MCP skill tools (`skill_search`, `skill_get`, `skill_validate`, `skill_get_reference`) are available when MCP is connected.

package/workflows/workflows/agent-environment-setup/platforms/gemini/skills/kaizen-iteration/SKILL.md

@@ -99,6 +99,6 @@ Provide a structured, metrics-driven framework for continuous improvement in sof
 - Gemini does not support `context: fork` — all skill execution is inline within the current session.
 - Skills are loaded via MCP when the Cubis Foundry MCP server is configured. Local `.agents/skills/` paths serve as hints.
 - User arguments are passed as natural language in the activation prompt.
-- Rules file: `.gemini/GEMINI.md`.
+- Rules file relative to the mirrored skill directory: `../../rules/GEMINI.md`.
 - Reference files are loaded relative to the skill directory under `.agents/skills/<skill-id>/`.
 - MCP skill tools (`skill_search`, `skill_get`, `skill_validate`, `skill_get_reference`) are available when MCP is connected.

package/workflows/workflows/agent-environment-setup/platforms/gemini/skills/kotlin-best-practices/SKILL.md

@@ -73,6 +73,6 @@ Produces Kotlin code using data classes, sealed hierarchies, structured coroutin
 - Gemini does not support `context: fork` — all skill execution is inline within the current session.
 - Skills are loaded via MCP when the Cubis Foundry MCP server is configured. Local `.agents/skills/` paths serve as hints.
 - User arguments are passed as natural language in the activation prompt.
-- Rules file: `.gemini/GEMINI.md`.
+- Rules file relative to the mirrored skill directory: `../../rules/GEMINI.md`.
 - Reference files are loaded relative to the skill directory under `.agents/skills/<skill-id>/`.
 - MCP skill tools (`skill_search`, `skill_get`, `skill_validate`, `skill_get_reference`) are available when MCP is connected.

package/workflows/workflows/agent-environment-setup/platforms/gemini/skills/kubernetes-deploy/SKILL.md

@@ -77,6 +77,6 @@ Provide complete Kubernetes manifests in YAML with inline comments explaining no
 - Gemini does not support `context: fork` — all skill execution is inline within the current session.
 - Skills are loaded via MCP when the Cubis Foundry MCP server is configured. Local `.agents/skills/` paths serve as hints.
 - User arguments are passed as natural language in the activation prompt.
-- Rules file: `.gemini/GEMINI.md`.
+- Rules file relative to the mirrored skill directory: `../../rules/GEMINI.md`.
 - Reference files are loaded relative to the skill directory under `.agents/skills/<skill-id>/`.
 - MCP skill tools (`skill_search`, `skill_get`, `skill_validate`, `skill_get_reference`) are available when MCP is connected.

package/workflows/workflows/agent-environment-setup/platforms/gemini/skills/mcp-server-builder/SKILL.md

@@ -1,6 +1,6 @@
 ---
 name: mcp-server-builder
-description: "Use when building MCP servers with spec-compliant tools, structured outputs, resource providers, transport configuration, and server testing."
+description: "Use when building MCP servers with spec-compliant tools, structured outputs, resource providers, transport configuration, and server testing. Anthropic compatibility aliases: mcp builder, mcp-builder."
 ---
 # MCP Server Builder
 
@@ -81,6 +81,11 @@ Guide the design and implementation of production-grade Model Context Protocol (
 - "Add a resource provider that serves database schema documentation to connected agents."
 - "Test an MCP server's tool registration and error handling with a mock client."
 
+## Compatibility Aliases
+
+Anthropic upstream compatibility names for this skill: `mcp builder`, `mcp-builder`.
+Treat requests that use those names as equivalent to this canonical Foundry skill.
+
 ## Gemini Platform Notes
 
 - Workflow and agent routes are compiled into `.gemini/commands/*.toml` TOML command files.
@@ -89,6 +94,6 @@ Guide the design and implementation of production-grade Model Context Protocol (
 - Gemini does not support `context: fork` — all skill execution is inline within the current session.
 - Skills are loaded via MCP when the Cubis Foundry MCP server is configured. Local `.agents/skills/` paths serve as hints.
 - User arguments are passed as natural language in the activation prompt.
-- Rules file: `.gemini/GEMINI.md`.
+- Rules file relative to the mirrored skill directory: `../../rules/GEMINI.md`.
 - Reference files are loaded relative to the skill directory under `.agents/skills/<skill-id>/`.
 - MCP skill tools (`skill_search`, `skill_get`, `skill_validate`, `skill_get_reference`) are available when MCP is connected.

package/workflows/workflows/agent-environment-setup/platforms/gemini/skills/nestjs/SKILL.md

@@ -73,6 +73,6 @@ Provide implementation code, module definitions, decorator usage, configuration
 - Gemini does not support `context: fork` — all skill execution is inline within the current session.
 - Skills are loaded via MCP when the Cubis Foundry MCP server is configured. Local `.agents/skills/` paths serve as hints.
 - User arguments are passed as natural language in the activation prompt.
-- Rules file: `.gemini/GEMINI.md`.
+- Rules file relative to the mirrored skill directory: `../../rules/GEMINI.md`.
 - Reference files are loaded relative to the skill directory under `.agents/skills/<skill-id>/`.
 - MCP skill tools (`skill_search`, `skill_get`, `skill_validate`, `skill_get_reference`) are available when MCP is connected.

package/workflows/workflows/agent-environment-setup/platforms/gemini/skills/nextjs/SKILL.md

@@ -86,6 +86,6 @@ Load detailed guidance based on context:
 - Gemini does not support `context: fork` — all skill execution is inline within the current session.
 - Skills are loaded via MCP when the Cubis Foundry MCP server is configured. Local `.agents/skills/` paths serve as hints.
 - User arguments are passed as natural language in the activation prompt.
-- Rules file: `.gemini/GEMINI.md`.
+- Rules file relative to the mirrored skill directory: `../../rules/GEMINI.md`.
 - Reference files are loaded relative to the skill directory under `.agents/skills/<skill-id>/`.
 - MCP skill tools (`skill_search`, `skill_get`, `skill_validate`, `skill_get_reference`) are available when MCP is connected.

package/workflows/workflows/agent-environment-setup/platforms/gemini/skills/observability/SKILL.md

@@ -77,6 +77,6 @@ Provide complete instrumentation code, configuration files, and alert rule defin
 - Gemini does not support `context: fork` — all skill execution is inline within the current session.
 - Skills are loaded via MCP when the Cubis Foundry MCP server is configured. Local `.agents/skills/` paths serve as hints.
 - User arguments are passed as natural language in the activation prompt.
-- Rules file: `.gemini/GEMINI.md`.
+- Rules file relative to the mirrored skill directory: `../../rules/GEMINI.md`.
 - Reference files are loaded relative to the skill directory under `.agents/skills/<skill-id>/`.
 - MCP skill tools (`skill_search`, `skill_get`, `skill_validate`, `skill_get_reference`) are available when MCP is connected.

package/workflows/workflows/agent-environment-setup/platforms/gemini/skills/owasp-security-review/SKILL.md

@@ -102,6 +102,6 @@ Perform structured, OWASP-standards-based security reviews of application code a
 - Gemini does not support `context: fork` — all skill execution is inline within the current session.
 - Skills are loaded via MCP when the Cubis Foundry MCP server is configured. Local `.agents/skills/` paths serve as hints.
 - User arguments are passed as natural language in the activation prompt.
-- Rules file: `.gemini/GEMINI.md`.
+- Rules file relative to the mirrored skill directory: `../../rules/GEMINI.md`.
 - Reference files are loaded relative to the skill directory under `.agents/skills/<skill-id>/`.
 - MCP skill tools (`skill_search`, `skill_get`, `skill_validate`, `skill_get_reference`) are available when MCP is connected.

package/workflows/workflows/agent-environment-setup/platforms/gemini/skills/pentest-skill/SKILL.md

@@ -103,6 +103,6 @@ Provide structured guidance for authorized penetration testing engagements follo
 - Gemini does not support `context: fork` — all skill execution is inline within the current session.
 - Skills are loaded via MCP when the Cubis Foundry MCP server is configured. Local `.agents/skills/` paths serve as hints.
 - User arguments are passed as natural language in the activation prompt.
-- Rules file: `.gemini/GEMINI.md`.
+- Rules file relative to the mirrored skill directory: `../../rules/GEMINI.md`.
 - Reference files are loaded relative to the skill directory under `.agents/skills/<skill-id>/`.
 - MCP skill tools (`skill_search`, `skill_get`, `skill_validate`, `skill_get_reference`) are available when MCP is connected.

package/workflows/workflows/agent-environment-setup/platforms/gemini/skills/playwright-interactive/SKILL.md

@@ -1,6 +1,6 @@
 ---
 name: playwright-interactive
-description: "Use when building interactive Playwright suites for browser automation, visual regression, accessibility audits, network interception, and collaborative test workflows."
+description: "Use when building interactive Playwright suites for browser automation, visual regression, accessibility audits, network interception, and collaborative test workflows. Anthropic compatibility aliases: webapp testing, webapp-testing."
 ---
 # Playwright Interactive
 
@@ -76,6 +76,11 @@ Deliver:
 | `agents/visual-reviewer.md` | Delegating screenshot comparison and visual regression tasks |
 | `agents/accessibility-auditor.md` | Delegating accessibility scans and report generation |
 
+## Compatibility Aliases
+
+Anthropic upstream compatibility names for this skill: `webapp testing`, `webapp-testing`.
+Treat requests that use those names as equivalent to this canonical Foundry skill.
+
 ## Gemini Platform Notes
 
 - Workflow and agent routes are compiled into `.gemini/commands/*.toml` TOML command files.
@@ -84,6 +89,6 @@ Deliver:
 - Gemini does not support `context: fork` — all skill execution is inline within the current session.
 - Skills are loaded via MCP when the Cubis Foundry MCP server is configured. Local `.agents/skills/` paths serve as hints.
 - User arguments are passed as natural language in the activation prompt.
-- Rules file: `.gemini/GEMINI.md`.
+- Rules file relative to the mirrored skill directory: `../../rules/GEMINI.md`.
 - Reference files are loaded relative to the skill directory under `.agents/skills/<skill-id>/`.
 - MCP skill tools (`skill_search`, `skill_get`, `skill_validate`, `skill_get_reference`) are available when MCP is connected.

package/workflows/workflows/agent-environment-setup/platforms/gemini/skills/playwright-web-qa/SKILL.md

@@ -0,0 +1,78 @@
+---
+name: playwright-web-qa
+description: "Use when running charter-driven web QA through Playwright MCP with deterministic evidence, runtime traces, and report-ready browser artifacts."
+---
+# Playwright Web QA
+
+## Purpose
+
+Provide a runtime-oriented browser QA workflow for validating real user journeys through Playwright MCP. This skill is for live charter execution and evidence review, not broad suite design.
+
+## When to Use
+
+- Running a manual or AI-authored QA charter against a real web app
+- Capturing screenshots, DOM state, console logs, network evidence, or accessibility evidence
+- Reproducing a bug with deterministic browser artifacts
+- Checking whether a page is blocked, broken, or visibly regressed before test-suite authoring
+
+## Instructions
+
+1. Confirm the Playwright MCP server is reachable before planning steps.
+2. Require a QA charter with `flow`, `success_criteria`, `base_url`, and `steps`.
+3. Start from an explicit URL state and avoid hidden navigation assumptions.
+4. Use Playwright MCP as the execution path instead of ad hoc browser scripts.
+5. Prefer stable selectors: role or test id first, then text, then CSS as a last resort.
+6. Capture baseline evidence before the first destructive action.
+7. Persist artifacts under `artifacts/web-qa/` with deterministic filenames.
+8. Capture console, network, or accessibility evidence only when the charter asks for it or when a failure needs explanation.
+9. Use one controlled retry only, then stop and report evidence.
+10. Tie the final verdict to artifact paths, not impressions.
+
+## Output Format
+
+```markdown
+## Web QA Charter
+- Flow:
+- Base URL:
+- Start path:
+- Success criteria:
+
+## Execution Log
+1. [action] -> [observation]
+2. [action] -> [observation]
+
+## Assertions
+- Passed:
+- Failed:
+- Blocked:
+
+## Evidence
+- Screenshots:
+- DOM snapshots:
+- Console logs:
+- Network evidence:
+- Accessibility evidence:
+
+## Risks And Follow-ups
+- Risk:
+- Recommended fix:
+```
+
+## References
+
+| File | Load when |
+| --- | --- |
+| `../playwright-interactive/SKILL.md` | The task shifts from runtime QA into suite authoring or broader E2E architecture. |
+| `../flutter-mobile-qa/SKILL.md` | The task also includes Android/mobile QA and needs the parallel mobile runtime pattern. |
+
+## Gemini Platform Notes
+
+- Workflow and agent routes are compiled into `.gemini/commands/*.toml` TOML command files.
+- Commands use `{{args}}` for user input, `!{shell command}` for shell output, `@{file}` for file content.
+- Specialists are internal postures (modes of reasoning), not spawned subagent processes.
+- Gemini does not support `context: fork` — all skill execution is inline within the current session.
+- Skills are loaded via MCP when the Cubis Foundry MCP server is configured. Local `.agents/skills/` paths serve as hints.
+- User arguments are passed as natural language in the activation prompt.
+- Rules file relative to the mirrored skill directory: `../../rules/GEMINI.md`.
+- Reference files are loaded relative to the skill directory under `.agents/skills/<skill-id>/`.
+- MCP skill tools (`skill_search`, `skill_get`, `skill_validate`, `skill_get_reference`) are available when MCP is connected.

package/workflows/workflows/agent-environment-setup/platforms/gemini/skills/prisma/SKILL.md

@@ -76,6 +76,6 @@ Load only what the current task requires.
 - Gemini does not support `context: fork` — all skill execution is inline within the current session.
 - Skills are loaded via MCP when the Cubis Foundry MCP server is configured. Local `.agents/skills/` paths serve as hints.
 - User arguments are passed as natural language in the activation prompt.
-- Rules file: `.gemini/GEMINI.md`.
+- Rules file relative to the mirrored skill directory: `../../rules/GEMINI.md`.
 - Reference files are loaded relative to the skill directory under `.agents/skills/<skill-id>/`.
 - MCP skill tools (`skill_search`, `skill_get`, `skill_validate`, `skill_get_reference`) are available when MCP is connected.

package/workflows/workflows/agent-environment-setup/platforms/gemini/skills/prompt-engineering/SKILL.md

@@ -79,6 +79,6 @@ Provide system prompts as formatted text blocks with labeled sections, few-shot
 - Gemini does not support `context: fork` — all skill execution is inline within the current session.
 - Skills are loaded via MCP when the Cubis Foundry MCP server is configured. Local `.agents/skills/` paths serve as hints.
 - User arguments are passed as natural language in the activation prompt.
-- Rules file: `.gemini/GEMINI.md`.
+- Rules file relative to the mirrored skill directory: `../../rules/GEMINI.md`.
 - Reference files are loaded relative to the skill directory under `.agents/skills/<skill-id>/`.
 - MCP skill tools (`skill_search`, `skill_get`, `skill_validate`, `skill_get_reference`) are available when MCP is connected.

package/workflows/workflows/agent-environment-setup/platforms/gemini/skills/python-best-practices/SKILL.md

@@ -82,6 +82,6 @@ No helper scripts are required for this skill right now. Keep execution in `SKIL
 - Gemini does not support `context: fork` — all skill execution is inline within the current session.
 - Skills are loaded via MCP when the Cubis Foundry MCP server is configured. Local `.agents/skills/` paths serve as hints.
 - User arguments are passed as natural language in the activation prompt.
-- Rules file: `.gemini/GEMINI.md`.
+- Rules file relative to the mirrored skill directory: `../../rules/GEMINI.md`.
 - Reference files are loaded relative to the skill directory under `.agents/skills/<skill-id>/`.
 - MCP skill tools (`skill_search`, `skill_get`, `skill_validate`, `skill_get_reference`) are available when MCP is connected.

package/workflows/workflows/agent-environment-setup/platforms/gemini/skills/react/SKILL.md

@@ -86,6 +86,6 @@ Load detailed guidance based on context:
 - Gemini does not support `context: fork` — all skill execution is inline within the current session.
 - Skills are loaded via MCP when the Cubis Foundry MCP server is configured. Local `.agents/skills/` paths serve as hints.
 - User arguments are passed as natural language in the activation prompt.
-- Rules file: `.gemini/GEMINI.md`.
+- Rules file relative to the mirrored skill directory: `../../rules/GEMINI.md`.
 - Reference files are loaded relative to the skill directory under `.agents/skills/<skill-id>/`.
 - MCP skill tools (`skill_search`, `skill_get`, `skill_validate`, `skill_get_reference`) are available when MCP is connected.

package/workflows/workflows/agent-environment-setup/platforms/gemini/skills/rust-best-practices/SKILL.md

@@ -82,6 +82,6 @@ No helper scripts are required for this skill right now. Keep execution in `SKIL
 - Gemini does not support `context: fork` — all skill execution is inline within the current session.
 - Skills are loaded via MCP when the Cubis Foundry MCP server is configured. Local `.agents/skills/` paths serve as hints.
 - User arguments are passed as natural language in the activation prompt.
-- Rules file: `.gemini/GEMINI.md`.
+- Rules file relative to the mirrored skill directory: `../../rules/GEMINI.md`.
 - Reference files are loaded relative to the skill directory under `.agents/skills/<skill-id>/`.
 - MCP skill tools (`skill_search`, `skill_get`, `skill_validate`, `skill_get_reference`) are available when MCP is connected.

package/workflows/workflows/agent-environment-setup/platforms/gemini/skills/secret-management/SKILL.md

@@ -98,6 +98,6 @@ Guide the design and implementation of robust secret and credential management s
 - Gemini does not support `context: fork` — all skill execution is inline within the current session.
 - Skills are loaded via MCP when the Cubis Foundry MCP server is configured. Local `.agents/skills/` paths serve as hints.
 - User arguments are passed as natural language in the activation prompt.
-- Rules file: `.gemini/GEMINI.md`.
+- Rules file relative to the mirrored skill directory: `../../rules/GEMINI.md`.
 - Reference files are loaded relative to the skill directory under `.agents/skills/<skill-id>/`.
 - MCP skill tools (`skill_search`, `skill_get`, `skill_validate`, `skill_get_reference`) are available when MCP is connected.

package/workflows/workflows/agent-environment-setup/platforms/gemini/skills/skill-creator/SKILL.md

@@ -161,6 +161,6 @@ skill-creator
 - Gemini does not support `context: fork` — all skill execution is inline within the current session.
 - Skills are loaded via MCP when the Cubis Foundry MCP server is configured. Local `.agents/skills/` paths serve as hints.
 - User arguments are passed as natural language in the activation prompt.
-- Rules file: `.gemini/GEMINI.md`.
+- Rules file relative to the mirrored skill directory: `../../rules/GEMINI.md`.
 - Reference files are loaded relative to the skill directory under `.agents/skills/<skill-id>/`.
 - MCP skill tools (`skill_search`, `skill_get`, `skill_validate`, `skill_get_reference`) are available when MCP is connected.

package/workflows/workflows/agent-environment-setup/platforms/gemini/skills/skill-creator/references/platform-adaptations.md

@@ -105,16 +105,30 @@ allowed-tools: [Read, Edit, Write, Bash, Grep, Glob]
 ## Adaptation Strategy
 
 ### Canonical Version (workflows/skills/)
-The canonical skill in `workflows/skills/<name>/` uses the full Pro Skill Standard format with rich frontmatter. This is the source of truth.
+The canonical skill in `workflows/skills/<name>/` is the authoring source of truth.
 
-### Claude Code Mirror
-Placed at `platforms/claude/skills/<name>/SKILL.md`. Uses full Claude frontmatter. Instructions may reference `context:fork`, `$ARGUMENTS`, and `${CLAUDE_SKILL_DIR}`.
+### Generated Platform Mirrors
+Generated mirrors are maintained for Claude Code, Copilot, Codex, Gemini CLI, and Antigravity under `platforms/<platform>/skills/<name>/SKILL.md`.
 
-### Copilot Mirror
-Placed at `platforms/copilot/skills/<name>/SKILL.md`. Same as Claude (Copilot reads Claude format). May add notes about prompt files and path-scoped instructions.
+- Claude Code mirrors can use Claude-specific frontmatter and instruction references such as `context: fork`, `$ARGUMENTS`, and `${CLAUDE_SKILL_DIR}`.
+- Copilot mirrors stay Claude-compatible where useful, but Copilot-native behavior still follows Copilot's own constraints.
+- Codex, Gemini CLI, and Antigravity mirrors usually keep minimal frontmatter and rely on platform rules plus MCP/routing behavior for the rest.
+- Do not assume one mirror transform works for every asset type. Skills, custom agents, workflow projections, prompts, and command files each have different platform constraints and must be generated through their own platform-aware pipeline.
 
-### Codex, Gemini CLI, Antigravity
-These platforms use the canonical version directly, referenced by their respective rule files. No separate mirror needed — the rule files handle routing.
+### Shared Steering + Platform Overrides
+Routing and loading behavior should not be documented independently in every platform skill.
+
+- `shared/rules/STEERING.md` is the canonical routing and skill-loading source.
+- `shared/rules/overrides/<platform>.md` adds platform-specific behavior.
+- Generated platform rule files are composed from the shared steering file plus the relevant platform override.
+
+### Non-Skill Assets
+Custom agents and workflow projections are not mirrored with the same logic as skills.
+
+- Skills are mirrored with platform-specific frontmatter and body transforms.
+- Shared agents are compiled into platform-native outputs such as Claude/Copilot markdown agents or Codex TOML agents.
+- Workflows are compiled into platform-native workflow skills, command files, or prompt files depending on the target platform.
+- Regenerate these with the platform asset generator, not by copying skill directories.
 
 ## Instruction Adaptation
 
@@ -124,5 +138,5 @@ When the canonical instructions reference platform-specific features, adapt them
 |-----------|------------|-------|--------|-------------|---------|
 | "fork to subagent" | `context: fork` | "switch posture" | "activate agent" | "use workflow" | `context: fork` |
 | "load reference" | `skill_get_reference` | "read file" | `activate_skill` | "read file" | `skill_get_reference` |
-| "use MCP tool" | `skill_validate` | N/A | `activate_skill` | N/A | N/A |
+| "use MCP tool" | `skill_validate` / `skill_get` | `skill_validate` / `skill_get` when MCP is connected | `skill_validate` / `skill_get` when MCP is connected; `activate_skill` only when using Gemini-native skill loading | `skill_validate` / `skill_get` when MCP is configured | `skill_validate` / `skill_get` when MCP is configured |
 | "run script" | `Bash` | `Bash` (may be sandboxed) | Shell | Shell | `Bash` |

package/workflows/workflows/agent-environment-setup/platforms/gemini/skills/skill-creator/references/routing-integration.md

@@ -18,15 +18,21 @@ User request
 
 ## Rule File Integration
 
-Each platform has a rule file with a skill routing matrix. When adding a new skill, update the matrix in every platform's rule file.
+Rule behavior is defined canonically in the shared steering layer, then specialized per platform by overrides and generated outputs.
+
+- Update `workflows/workflows/agent-environment-setup/shared/rules/STEERING.md` when the routing contract itself changes.
+- Update `workflows/workflows/agent-environment-setup/shared/rules/overrides/<platform>.md` only for platform-specific deviations.
+- Regenerate platform rule files after canonical edits instead of hand-editing every generated platform file.
+
+Each generated platform rule file still exposes a routing matrix or equivalent skill-routing surface, but the source of truth is the shared steering + override layer.
 
 ### Skill Routing Matrix Format
 
 ```markdown
 | Skill | Category | When to Load | Primary Agent |
 |-------|----------|--------------|---------------|
-| python-best-practices | Language | Python backend, typing, async | @backend-specialist |
-| nextjs | Framework | Next.js App Router, RSC | @frontend-specialist |
+| python-best-practices | Language | Python backend, typing, async | @implementer |
+| nextjs | Framework | Next.js App Router, RSC | @implementer |
 ```
 
 ### Matrix Fields
@@ -50,7 +56,9 @@ Each platform has a rule file with a skill routing matrix. When adding a new ski
 
 ### When to Create Mirrors
 
-Create platform-specific mirrors for Claude Code and Copilot. Other platforms use the canonical version.
+Create or regenerate platform skill mirrors for every supported platform. The authoring source remains canonical under `workflows/skills/`, but the install/runtime surfaces use generated platform mirrors.
+
+This applies to skills only. Do not reuse the skill mirror process for shared agents, workflow markdown, prompt files, or command files.
 
 ### Claude Code Mirror
 
@@ -77,7 +85,18 @@ Same format as Claude (Copilot reads Claude format). Add notes about:
 
 ### Codex / Gemini / Antigravity
 
-No mirror needed. The canonical version is used directly. Routing is handled by the platform rule file.
+These platforms also receive generated skill mirrors. Keep their frontmatter minimal and let the generated rule files plus MCP/routing behavior carry the platform-specific execution model.
+
+## Agent and Workflow Projections
+
+Shared agents and workflows must go through platform-specific generators because their output format differs by platform.
+
+- Codex agents are emitted as `.toml` native agents.
+- Claude and Copilot agents are emitted as markdown with sanitized platform frontmatter.
+- Gemini and Antigravity workflows and agent routes are emitted as TOML command files.
+- Copilot workflows are emitted as prompt files rather than skill mirrors.
+
+If a change affects custom agents, workflow prompts, or command behavior, update the shared source and rerun the platform asset generator instead of relying on `sync-skill-mirrors`.
 
 ## Agent Assignment
 
@@ -85,16 +104,16 @@ Each skill should be assigned a primary agent in the routing matrix. Choose base
 
 | Domain | Primary Agent |
 |--------|---------------|
-| Languages (Python, TS, Go, etc.) | @backend-specialist or @frontend-specialist |
-| Backend frameworks | @backend-specialist |
-| Frontend frameworks | @frontend-specialist |
-| Mobile frameworks | @mobile-developer |
-| Database/ORM | @database-architect |
-| Testing | @test-engineer |
-| Security | @security-auditor |
-| DevOps | @devops-engineer |
-| Design/Architecture | @project-planner |
-| AI/ML | @researcher |
+| Languages (Python, TS, Go, etc.) | @implementer |
+| Backend frameworks | @implementer |
+| Frontend frameworks | @implementer |
+| Mobile frameworks | @implementer |
+| Database/ORM | @planner or @implementer |
+| Testing | @tester |
+| Security | @reviewer |
+| DevOps | @implementer |
+| Design/Architecture | @planner |
+| AI/ML / research | @explorer or @planner |
 | Workflow | @orchestrator |
 | Meta | @orchestrator |
 
@@ -102,17 +121,20 @@ Each skill should be assigned a primary agent in the routing matrix. Choose base
 
 After adding a skill to the routing infrastructure:
 
-1. Run `npm run test:attributes` — validates frontmatter across all skills
-2. Run `npm run check:generated-assets` — verifies platform mirrors are in sync
-3. Verify the skill appears in `npm run generate:skills-index`
-4. Test `route_resolve` with a prompt that should match the new skill
-5. Test `skill_validate` with the new skill's name
-6. Test `skill_get` to confirm the SKILL.md loads correctly
+1. Update canonical skill content under `workflows/skills/<name>/`
+2. Update shared steering or platform overrides if the routing contract changed
+3. Run `npm run generate:all` or the narrower mirror/rule/platform-asset generation commands
+4. Run `npm run test:attributes` — validates frontmatter across all skills
+5. Run `npm run check:generated-assets` — verifies platform mirrors are in sync
+6. Verify the skill appears in `npm run generate:skills-index`
+7. Test `route_resolve` with a prompt that should match the new skill
+8. Test `skill_validate` with the new skill's name
+9. Test `skill_get` to confirm the SKILL.md loads correctly
 
 ## Common Integration Mistakes
 
-1. **Forgetting to update all 5 platform rule files** — every platform needs the skill in its routing matrix
+1. **Editing generated platform rule files directly** — change shared `STEERING.md` or a platform override, then regenerate
 2. **Mismatched names** — directory name, frontmatter name, and routing matrix name must all match
-3. **Wrong agent assignment** — assigning a frontend skill to @backend-specialist causes incorrect context loading
-4. **Missing mirror creation** — Claude and Copilot platforms need mirrors with platform-specific frontmatter
+3. **Wrong agent assignment** — assigning a testing or review skill to the wrong lean agent causes incorrect routing and validation
+4. **Using skill mirror sync for non-skill assets** — agents, prompts, and command files have their own platform-native generation path
 5. **Invalid evals.json** — JSON syntax errors cause the eval runner to skip the skill silently